Artificial intelligence-driven monitoring and sustainable management of lithium-ion battery waste: integrating advanced sensing, environmental remediation, and circular resource recovery.

The rapid growth of lithium-ion batteries (LIBs) is generating heterogeneous end-of-life streams that combine substantial resource value with fire, fluorochemical, and metal-contamination risks. This review critically examines how advanced sensing, artificial intelligence (AI), recycling chemistry, environmental remediation, and circular-resource strategies can be integrated across LIB waste management. Gas, thermal, electrochemical, spectroscopic, and imaging technologies are evaluated according to their monitoring targets and, where available, quantitative performance, with particular attention to HF and other fluorinated decomposition products. Evidence from primary AI studies is assessed using dataset origin and size, input variables, prediction targets, validation procedures, and reported performance rather than algorithm names alone. Hydrometallurgical recycling is examined mechanistically through reductive leaching, organic-acid chemistry, solvent extraction, and product recovery while distinguishing leaching efficiency from final material recovery and purity. Pyrometallurgy, direct recycling, bioleaching, and emerging solvent systems are compared without assuming a universally superior route. Importantly, recycling is distinguished from remediation of contaminants released to process water, soil, and gaseous streams. The resulting framework connects hazard detection, chemistry-aware sorting, data-driven process optimization, contaminant control, resource recovery, and digital traceability. Major limitations remain insufficient external validation of AI models, mixed-stream sensor calibration, inconsistent recovery metrics, feedstock variability, and limited industrial-scale evidence. Addressing these constraints is essential for developing genuinely safe, adaptive, and circular LIB waste-management systems.